INDEX

Explanations

entities related to different types of professions or roles within specific contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to various groups of people affected by social or economic issues

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 charm

-0.66

 fiasco

-0.65

 indictment

-0.62

 rumors

-0.61

 rumours

-0.60

bernatorial

-0.60

 rumor

-0.59

 wonders

-0.56

 plea

-0.56

 sunshine

-0.55

POSITIVE LOGITS

 alike

1.06

 themselves

0.99

'.

0.96

who

0.94

pace

0.93

hip

0.92

folk

0.90

paces

0.88

 irrespective

0.86

Activations Density 0.374%

entities related to different types of professions or roles within specific contexts

terms related to various groups of people affected by social or economic issues

No Comments

No Known Activations